MBS: a model for risk benefit analysis of breast cancer screening

The influence of disease risk on the optimal time interval between screens for the early detection of cancer: a mathematical approach

The intervals between screens for the early detection of diseases such as breast and colon cancer suggested by screening guidelines are typically based on the average population risk of disease. With the emergence of ever more biomarkers for cancer risk prediction and the development of personalized medicine, there is a need for risk-specific screening intervals. The interval between successive screens should be shorter with increasing cancer risk. A risk-dependent optimal interval is ideally derived from a cost-effectiveness analysis using a validated simulation model. However, this is time-consuming and costly. We propose a simplified mathematical approach for the exploratory analysis of the implications of risk level on optimal screening interval. We develop a mathematical model of the optimal screening interval for breast cancer screening. We verified the results by programming the simplified model in the MISCAN-Breast microsimulation model and comparing the results. We validated the results by comparing them with the results of a full, published MISCAN-Breast cost-effectiveness model for a number of different risk levels. The results of both the verification and validation were satisfactory. We conclude that the mathematical approach can indicate the impact of disease risk on the optimal screening interval.

Calibration methods used in cancer simulation models and suggested reporting guidelines

Increasingly, computer simulation models are used for economic and policy evaluation in cancer prevention and control. A model's predictions of key outcomes, such as screening effectiveness, depend on the values of unobservable natural history parameters. Calibration is the process of determining the values of unobservable parameters by constraining model output to replicate observed data. Because there are many approaches for model calibration and little consensus on best practices, we surveyed the literature to catalogue the use and reporting of these methods in cancer simulation models. We conducted a MEDLINE search (1980 through 2006) for articles on cancer-screening models and supplemented search results with articles from our personal reference databases. For each article, two authors independently abstracted pre-determined items using a standard form. Data items included cancer site, model type, methods used for determination of unobservable parameter values and description of any calibration protocol. All authors reached consensus on items of disagreement. Reviews and non-cancer models were excluded. Articles describing analytical models, which estimate parameters with statistical approaches (e.g. maximum likelihood) were catalogued separately. Models that included unobservable parameters were analysed and classified by whether calibration methods were reported and if so, the methods used. The review process yielded 154 articles that met our inclusion criteria and, of these, we concluded that 131 may have used calibration methods to determine model parameters. Although the term 'calibration' was not always used, descriptions of calibration or 'model fitting' were found in 50% (n = 66) of the articles, with an additional 16% (n = 21) providing a reference to methods. Calibration target data were identified in nearly all of these articles. Other methodological details, such as the goodness-of-fit metric, were discussed in 54% (n = 47 of 87) of the articles reporting calibration methods, while few details were provided on the algorithms used to search the parameter space. Our review shows that the use of cancer simulation modelling is increasing, although thorough descriptions of calibration procedures are rare in the published literature for these models. Calibration is a key component of model development and is central to the validity and credibility of subsequent analyses and inferences drawn from model predictions. To aid peer-review and facilitate discussion of modelling methods, we propose a standardized Calibration Reporting Checklist for model documentation.

Risk assessment in regulatory policy making for human and veterinary public health

Risk assessment is the method of systematically identifying and assessing factors that influence the probability and consequences of a negative event occurring. One responsibility of veterinary medicine is to protect animal and human health. Food animal production uses antibiotics to enhance production. Regulators evaluate new production technology to ensure animal safety and safe, edible products and to make public policy decisions by assessing risks/benefits. The U.S. Food and Drug Administration, Center for Veterinary Medicine's (CVM's) first risk assessment addressed the potential human health impact of campylobacter effects associated with the use of fluoroquinolines in food-producing animals. CVM used the Monte Carlo method to estimate risk byprobability distributions that reflect the uncertainty and variability in the data used for the assessment. Enterococci faecium is a species more likely to be resistant to antibiotics of last resort. Effective control of multidrug-resistant enterococci will requirea better understanding of the transfer of E. faeciumfrom animals to humans and the interaction between E. faecium, the hospital environment, and humans; prudent antibiotic use; better contact isolation in hospitals; and better surveillance. CVM will model these factors in a second, more complex risk assessment designed to examine the indirect transfer of resistance from animals to humans. Use of risk assessments allows researchers, the industry, regulatory authorities, and educators to make better policy decisions regarding antimicrobial use in food animals and humans and the development of resistance. Today the question of whether the use of antimicrobials for growth enhancement infood animals should or should not be terminated for the benefit of human health remains unresolved.

A mathematical algorithm that computes breast cancer sizes and doubling times detected by screening

This paper presents a mathematical algorithm that computes the sizes and growth rates of breast cancer detected in a hypothetical population that is screened for the disease. The algorithm works by simulating the outcomes of the hypothetical population twice, first without screening and then with screening. The simulation without screening relies on an underlying model of the natural history of the disease. The simulation with screening uses this natural history model to track the growth of breast tumors backwards in the time starting from the time they would have been detected without screening. The method of tracking tumor growth backward in time is different from methods that track tumor growth forward in time by starting from an estimated time of tumor onset. The screening algorithm combines the natural history model, the method tracking of tumor growth backward in time, the age group, the interval between screening exams, and the detection threshold of the screening exam to compute the joint distribution of tumor size and growth rate among screen-detected and interval patients. The algorithm also computes the sensitivity and leadtime distribution. It allows for arbitrary age groups, detection thresholds and screening intervals and may contribute to the design of future screening trials.

The effects of fractionated gamma irradiation on induction of mammary carcinoma in normal and estrogen-treated rats

The effects of dose fractionation on induction of mammary carcinoma were studied in normal and estrogen-treated female rats of the inbred WAG/Rij strain. Groups of 40 animals received total-body doses of 1 or 2 Gy of (137)Cs gamma radiation, administered in fractions of 2.5, 10 or 40 mGy with intervals of 12 h, or in fractions of 10 mGy with intervals of 2, 5 or 24 h. The irradiations were started at the age of 8 weeks. Estrogen treatment was accomplished by implantation of a pellet containing estrogen at the age of 6 weeks. All mammary tumors were resected and classified histologically as carcinoma or fibroadenoma. The age-specific incidence of mammary carcinoma was compared with that in control groups of unirradiated normal or estrogen-treated rats and was expressed as excess normalized risk, using lifetime statistical analysis with both parametric and nonparametric methods. The data were also compared to the results of single-dose experiments reported in previous papers. Fractionated irradiation increased the risk of mammary cancer in both normal and estrogen-treated rats compared to the corresponding unirradiated control group. The excess normalized risk per unit of total dose was approximately equal with or without estrogen treatment. Without estrogen treatment, the effects of the single-dose and fractionated irradiations were approximately equal. In estrogen-treated animals, however, single-dose irradiation was up to 15 times more carcinogenic than the fractionated exposures. This fractionation effect appeared to vanish for total doses below approximately 0.3 Gy. With estrogen treatment, the excess normalized risk was significantly higher for dose fractions of 40 mGy than for fractions of 10 mGy. The risk was also markedly higher for fractionation intervals of 2 or 5 h than for intervals of 12 or 24 h. The results of these experiments show that the effects of dose fractionation on the induction of mammary carcinoma may depend on hormonal status, the total dose delivered, the dose per fraction, and the fractionation interval.

Use of a mathematical model to evaluate breast cancer screening policy

A model of breast cancer screening was developed, in which the processes of tumour origination and growth, detection of tumours at screening, presentation of women with cancers to their GP, and of survival after diagnosis were modelled parametrically. The model was fitted to data from the North-West of the UK, for 413 women who screened positive, and for 761 women who developed interval cancers. Model validation comprised verification that the final model fitted the data adequately, together with the comparison of model predictions with findings by other workers. The mathematical model was used to assess different screening policies, and to ask "what if" questions. Taking the cost of breast cancer to be the sum of the cost of screening and the cost of PYLL (person years of life lost due to cancer), the optimal screening policy was calculated. The costs of the current policy and of other possible screening policies were found, together with their effects on life lost and on mortality. The tentative conclusion was that if monies can be found to extend the screening programme, for example to carry out one more screen per woman, most benefit would be obtained by reducing the start age of screening by 3 years.

Assessment of lifetime gained as a result of mammographic breast cancer screening using a computer model

A computer model for the simulation of breast cancer screening (MBS) is used to calculate the results of screening in terms of lifetime. To optimize breast cancer screening protocols, risk (lifetime lost due to radiation-induced tumours) versus benefit (lifetime gained due to early detection of breast cancer) analyses are performed for a simulated stable Swedish female population. The present study focuses on the results of different screening strategies employing single view mammography, including starting and finishing ages of screening, time interval between successive screening sessions as well as the influence of high detection screening and differences between different populations, based on lifetime lost or gained. To test the stability of the recommendations with respect to possible changes in the variables used in MBS, calculations are also performed for high risk factors for breast tumour induction using both the additive and multiplicative risk models, fast growing breast tumours, late incidence of breast tumours and age dependent survival. The results of the simulations expressed in terms of lifetime gained suggest that a theoretical benefit can be obtained employing starting and finishing ages of 35 and 75 years, respectively. In terms of number of fatal breast tumours, the favourable screening period is 40-80 years. It is concluded that the recommendations are stable for changes in the input variables of MBS. The benefits of higher detection screening are more marked for younger than for older women. A high screening frequency results in more lifetime gained, especially at relatively young ages, whereas for older ages the effect is only marginal.

Optimisation of mammographic breast cancer screening using a computer simulation model

To optimise breast cancer screening protocols, risk (induction of fatal tumors) versus benefit (reduction in the number of fatal tumors) analyses are performed for a simulated stable Swedish female population, using the Model for evaluation of Breast cancer Screening (MBS). The present study comprises, the influences of various screening parameters, i.e. ages at which screening is started and stopped, interval period between successive sessions, tumor detection limits for screening and average glandular dose per screening session. When the results of the present study are expressed in terms of numbers of fatal breast tumors, it appears that starting and stopping ages for screening of 40 and 80 years, respectively, seem realistic. An increased screening frequency results in a larger reduction of breast cancer mortality. This reduction is significant for ages between 40 and 51 years but only marginal for ages above 70 years. High resolution screening, i.e., the detection of tumors at smaller size, results in a larger benefit but does not indicate a younger age for starting of screening. The average glandular dose per screening session does only influence the risks of screening. As separate risk and benefit results are presented, a change in average glandular dose on the total effect of screening can easily be calculated.